The devices, systems, and methods described herein enable automatically configuring an electronic device using artificial intelligence (AI). The devices, systems, and methods enable accessing telemetry data representing device usage data, inputting the accessed telemetry data into machine learning models that are matched to device metadata, and determining notifications to publish to components of the electronic device. The notifications represent events predicted to occur on the electronic device. The notifications are published to the components of the electronic device such that the electronic device is configured according to the published notifications. The determined notifications enable the identification of optimal settings for the electronic device based on the usage pattern of the device and enable components of the electronic device to preemptively take action on events which are predicted to occur in the future.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
2. The electronic device of claim 1, wherein the component includes an operating system of the electronic device and the client service has computer-executable instructions that, in response to execution by the processor, determine a notification relating to at least one setting of the operating system, and wherein configuring, by the component, includes configuring the at least one setting of the operating system based at least on the published notification.
3. The electronic device of claim 1, wherein the component includes at least one of an application of the electronic device or hardware of the electronic device, and the client service has computer-executable instructions that, in response to execution by the processor, determine a notification relating to the at least one of the application or the hardware, and wherein configuring, by the component, includes configuring at least one of the application or the hardware based at least on the published notification.
4. The electronic device of claim 1, wherein the information that is associated with at least one of the electronic device or the user of the electronic device and used to update the received at least one machine learning model is unknown to the cloud service.
This invention relates to privacy-preserving machine learning in electronic devices, addressing the challenge of improving local machine learning models without exposing sensitive user or device data to cloud services. The system involves an electronic device that receives at least one machine learning model from a cloud service. The device then updates this model using locally available information associated with the device or its user. Critically, the information used for these updates remains unknown to the cloud service, ensuring privacy. The device may also receive additional machine learning models from the cloud service, which are similarly updated locally without revealing the underlying data. The system may further include a user interface for managing model updates and privacy settings. The invention ensures that while machine learning models are enhanced through local data, the sensitive information used for these updates is never transmitted to or accessible by the cloud service, addressing concerns about data privacy and security. This approach allows for personalized model improvements while maintaining confidentiality of user and device-specific information.
5. The electronic device of claim 1, wherein the component includes hardware of the electronic device and the client service has computer-executable instructions that, in response to execution by the processor, determine a notification relating to the hardware, the notification representing a predicted usage of the hardware based at least on the telemetry data, and wherein configuring, by the component, includes configuring the hardware with at least one application relating to the predicted usage of the hardware.
6. The electronic device of claim 1, further comprising a universal telemetry client that has computer-executable instructions that, in response to execution by the processor, upload the telemetry data from the memory to the cloud service for updating the at least one machine learning model at the cloud service based on at least one of the telemetry data of the electronic device or telemetry data from another electronic device.
7. The electronic device of claim 1, wherein the telemetry data input by the client service from the memory into the at least one machine learning model includes real-time telemetry data from the electronic device.
8. The electronic device of claim 1, wherein the client service has computer-executable instructions that, in response to execution by the processor, reject the determined notification based on information that is associated with at least one of the user or the electronic device.
9. The electronic device of claim 1, wherein the information that is associated with at least one of the electronic device or the user of the electronic device and used to update the at least one machine learning model comprises at least one of custom hardware of the electronic device, a proprietary chip, hardware of the electronic device, software of the electronic device, whether the electronic device is a work device, or whether the electronic device is a personal device.
This invention relates to electronic devices that use machine learning models to enhance functionality, particularly in distinguishing between work and personal device usage. The core challenge addressed is improving the accuracy and adaptability of machine learning models by incorporating device-specific and user-specific data. The electronic device includes at least one machine learning model that is updated based on information associated with the device or its user. This information may include custom hardware, proprietary chips, general hardware or software components of the device, or whether the device is designated as a work or personal device. By leveraging these factors, the machine learning model can better tailor its operations to the device's context, improving performance and user experience. The system dynamically adjusts the model based on the device's configuration and usage patterns, ensuring more personalized and efficient functionality. This approach enhances security, productivity, and user satisfaction by aligning the device's behavior with its intended use case, whether for professional or personal purposes. The invention aims to provide a more intelligent and responsive electronic device by continuously refining the machine learning model with relevant device and user data.
10. The electronic device of claim 1, wherein the client service has computer-executable instructions that, in response to execution by the at least one processor, request the at least one machine learning model from the cloud service based on the device metadata, the at least one machine learning model being an open neural network exchange (ONNX) model.
12. The method of claim 11, wherein inputting the accessed telemetry data into the updated at least one machine learning model includes determining a notification relating to at least one setting of an operating system of the electronic device, and wherein configuring, by the component, includes configuring the at least one setting of the operating system based at least on the published notification.
13. The device implemented method of claim 11, wherein inputting the accessed telemetry data into the updated at least one machine learning model includes determining a notification relating to at least one of an application of the electronic device or hardware of the electronic device, and wherein configuring, by the component, includes configuring the at least one of the application or the hardware based at least on the published notification.
This invention relates to a method for processing telemetry data from an electronic device using machine learning to generate and apply notifications for optimizing device performance. The method involves accessing telemetry data from the device, which may include usage patterns, performance metrics, or error logs. This data is input into a machine learning model that has been updated based on prior telemetry data and user feedback. The model analyzes the data to determine notifications related to either an application running on the device or the device's hardware. These notifications may indicate performance issues, optimization opportunities, or required updates. The method then configures the application or hardware based on the generated notification, such as adjusting settings, applying patches, or reallocating resources. The goal is to improve device functionality, efficiency, or user experience by leveraging predictive insights from machine learning. The system may also publish these notifications to a central repository for broader analysis or distribution. This approach automates device management by using real-time data and adaptive learning to proactively address potential issues or enhancements.
14. The method of claim 11, wherein the information that is associated with at least one of the electronic device or the user of the electronic device and used to update the received at least one machine learning model is unknown to the cloud service.
15. The method of claim 11, wherein inputting the accessed telemetry data into the updated at least one machine learning model includes determining a notification relating to hardware of the electronic device, the notification representing a predicted usage of the hardware based at least on the telemetry data, and wherein configuring, by the component, includes configuring the hardware with at least one application relating to the predicted usage of the hardware.
17. The one or more computer storage media of claim 16, wherein the processor is configured to input the accessed telemetry data into the updated at least one machine learning model to determine a notification relating to at least one of (i) at least one setting of an operating system of the electronic device, (ii) an application of the electronic device, or (iii) hardware of the electronic device, and wherein configuring, by the component, includes configuring the at least one of (i) the at least one setting of the operating system, (ii) the application, or (iii) the hardware based on the published notification.
18. The one or more computer storage media of claim 16, wherein the at least one machine learning model matched to the device metadata relates to an operating system of the electronic device.
19. The one or more computer storage media of claim 16, wherein the processor is configured to input the accessed telemetry data into the updated at least one machine learning model to determine a notification relating to hardware of the electronic device, the notification representing a predicted usage of the hardware based on the telemetry data, and wherein configuring, by the component, includes configuring the hardware with at least one application relating to the predicted usage of the hardware.
This invention relates to predictive hardware management in electronic devices using machine learning. The system collects telemetry data from an electronic device, such as usage patterns, performance metrics, and environmental conditions. A machine learning model processes this data to predict future hardware usage, such as component wear, performance degradation, or optimal configurations. Based on these predictions, the system generates notifications or alerts about the hardware's expected behavior. The system then configures the hardware with applications or settings tailored to the predicted usage, optimizing performance and longevity. For example, if the model predicts increased CPU demand, the system may preload relevant applications or adjust power settings. The machine learning model is periodically updated with new telemetry data to improve accuracy. The system ensures efficient hardware utilization by proactively adapting to anticipated usage patterns, reducing downtime and enhancing device performance.
20. The one or more computer storage media of claim 16, wherein the processor is further configured to upload the telemetry data to the cloud service for updating the at least one machine learning model at the cloud service based on at least one of the telemetry data of the electronic device or telemetry data from another electronic device.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
May 31, 2018
November 8, 2022
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.